Synopsis On the morning of 13 August 2002, the commercial salmon fishing vessel CapRougeII was bound for the entrance of the main arm of the Fraser River, British Columbia. When the vessel was approximately two miles south of Sand Heads Light, it capsized with seven persons on board. Two persons abandoned the vessel and climbed into a skiff being towed by the fishing vessel. Five persons, including two children, remained within the overturned hull and drowned. Ce rapport est galement disponible en franais. 1.0 Factual Information 1.1 Particulars of the Vessel 1.1.1 Vessel Description The CapRougeII was a small fishing vessel of closed construction, having an all-welded aluminum single-chine hull with a transom stern, incorporating an hydraulic fish loading ramp which was generally maintained in the raised position. The hull below the main deck was subdivided by transverse watertight bulkheads, enclosing (from forward) the crew accommodation space, engine room, two forward and two after insulated fish holds divided by a centreline bulkhead, and a lazarette in which the hydraulic steering gear was located. Built-in diesel tanks were arranged on the port and starboard sides of the engine room and the lazarette, and a fresh water storage tank was located under the crew accommodation forward of the engine room. The vessel was powered by a marine diesel engine with a reverse/reduction gear box driving a single fixed-pitch propeller and was fitted with a single plate centreline rudder (see Figure1). The vessel was generally engaged in seine fishing operations and was equipped with an hydraulic purse seining winch and net storage drum. The main mast, located near midships, was rigged with a cargo boom, fitted with hydraulic topping lift and vanging winches. The mast was also fitted with a crosstree and rigging, providing for the vertical stowage and deployment of a pair of outrigger booms, equipped with roll reduction paravanes (often referred to as stabilizers). An aluminum superstructure enclosed the wheelhouse, captain's cabin, galley, mess, washroom, engine room access, and escape hatch. A weathertight hinged door, giving access to the forecastle deck, was located on the starboard side of the wheelhouse. A combined hinged hatch cover and door on the centreline at the after end of the wheelhouse provided an alternative exit and also gave access to the outside steering position, located on top of the main deckhouse (see Photo8). A weathertight hinged door, giving access from the galley to the after end of the main deck, was located to starboard of the centreline in the after end bulkhead of the deckhouse. The raised wheelhouse was arranged with a comprehensive outfit of communications and navigation equipment, including radar, VHFradio, depth sounder, compass, autopilot, navigation lights and sound signals control panels. Figure 1. Outline general arrangement 1.2 Fishery Opening At approximately 13002 on Friday 09August 2002, the Department of Fisheries and Oceans (DFO) announced that it would permit 32licenced commercial seine fishing vessels to fish for sockeye salmon in a designated portion of the Strait of Juan de Fuca south of Port Renfrew, British Columbia, known as Fisheries Statistical Area 20 (see Figure2). The fishery would commence at 0600 on Monday, August12, and end eight hours later at 1400. Under the terms of a joint venture agreement between DFO and private industry, the exact identity of the 32vessels would be decided by private industry. Regardless of which vessels were chosen, DFO required each of them to be on the fishing grounds no later than 2000 on the evening of August11. Vessels scheduled to participate in the Area20 fishery organized themselves into catch-sharing groups, usually consisting of between six and eight vessels. The total number of fish caught by all the vessels participating in any one group would be divided equally amongst all its members. Originally, the CapRougeII was meant to participate in a salmon fishery scheduled to take place at the same time as the Area20 fishery, but in a more geographically protected waterway known as Johnstone Strait, considered by DFO to be part of Fisheries Statistical Area12. When a seiner, originally chosen by the catch-sharing group to participate in the Area20 fishery, had insufficient time to provision itself and reach Port Renfrew by the DFO-required time on August 11, CapRougeII was called upon to forego its plans to participate in Area12 and was re-directed to Area20. Fishing nets permitted in Area 12 (inside seine nets) were shorter, shallower and usually weighed approximately 2500kg less than those permitted in Area20. Figure 2. Commercial salmon seine areas 1.3 History of the Voyage To prepare for its participation, the CapRougeII, with a total of five crew including the skipper3, left Sturdies Bay on Galiano Island on the morning of August10 and made its way across the Strait of Georgia to Steveston, a community within the City of Richmond, located on the shores of the south arm of the Fraser River. In Steveston, the crew acquired groceries, fuel, ice for the fish holds, and fishing gear which included a west coast seine net. Later in the day, the CapRougeII left Steveston for Whaler Bay, on the east side of Galiano Island. There, the CapRougeII took in tow a diesel-powered aluminum seine skiff and continued its voyage as far as Cadboro Bay near Victoria, where it anchored for the night. The next morning at approximately 0630, the vessel weighed anchor and continued its voyage to Port Renfrew in the Strait of Juan de Fuca, arriving there at approximately 1315. At 0600 on August 12, the CapRougeII, along with the other 31vessels, began fishing. The fishing continued for eight hours until 1400. After the skipper reported to DFO the quantity of his catch, he took on board all the fish caught by another vessel participating in the same group. The CapRougeII had approximately 22850pounds of fish on board when it left the fishing grounds at approximately 1845. This quantity was about half of its cargo capacity. The fish were stowed in three of four insulated holds (see Figure 1). Each hold was designed to contain a slurry of seawater and ice chips in which the fish were carried to the processing facility. All except approximately 200pounds of fish were evenly distributed between the two forward holds. The remaining 200pounds of fish were stowed in the port after hold and none in the starboard after hold. Each of the four holds was partially filled with seawater. The two forward holds were approximately 2/3 to 3/4 full and the after holds were approximately full. In preparation for the voyage, a painter was attached to a bridle connected directly to the port and starboard bows of the power skiff and led forward over the stern roller of the fish loading ramp, forward over the top of the seine net wound on the drum and secured to the purse seining winch, located forward of the fish hold hatch. The skiff was initially cinched tight against the roller of the stern ramp of the CapRougeII. (Subsequently, the bow of the skiff came free of the stern roller as the painter stretched while the vessel was underway.) This work was done while the CapRougeII was stationary, prior to the commencement of its voyage to the processor. At approximately 1845, the CapRougeII departed the Port Renfrew area in clear, calm weather. The lazarette was pumped free of sea water some four to five hours later. The next morning, the vessel arrived back in Sturdies Bay so that two children, whose parents were already on board the vessel, could travel as guests for the remainder of the voyage to Steveston. At approximately 0700 on August13, the vessel departed Sturdies Bay and made its way into the Strait of Georgia toward the entrance of the north arm of the Fraser River. Environment Canada reports that, at the time, the weather was clear with 22knot winds blowing from the northwest. Wave height was reported as being between one and two metres. The vessel was on a northerly course proceeding at approximately six knots and steered by automatic pilot. At 0857, the vessel made a starboard to starboard passing with the 33.5m sailing vessel FrancisLynnII, which was under full sail and approximately four miles outbound for Active Pass from Sand Heads at the entrance to the Fraser River. The vessels passed within 180 m of each other. The mate on board the sailing vessel was on watch at the exterior helm near the stern of the sailing vessel. He recalled that, at the time of the vessels' passing, the CapRougeII was trimmed heavily by the stern and seawater was washing onto its main deck. On board the CapRougeII, two crew members were on watch while a third crew member was in the wheelhouse but not officially on watch. The skipper was sleeping in his cabin on the starboard side of the main deck while a fourth crew member was sleeping on a settee located on the port side of the galley. The boy, who had joined the vessel in Sturdies Bay, was in the wheelhouse with the three adults. His sister was resting in one of the bunks, located in the forward crew accommodation area below decks. One of the crew members woke the skipper to inform him that the seiner had a starboard list which he might want to assess and correct. In response, the skipper left his cabin and made his way aft to the galley door, which was secured in the open position. This door led to the main working deck. He observed seawater on the starboard side of the deck, but due to the obstruction caused by the seine net wrapped on the net drum, he was unable to see the stern of the vessel or the skiff. The skipper descended a ladder into the engine room and started an auxiliary engine used to power the vessel's bilge pumps. He then proceeded aft along the starboard side of the engine. To correct the list, he adjusted the valves necessary to pump overboard water in the starboard after fish hold. Pumping had just begun when the vessel suddenly heeled heavily to starboard, causing unsecured equipment to fall from the top of the port side main fuel tank. The skipper made his way up the ladder into the interior companionway of the main deck and shouted to those in the wheelhouse to take the engine out of gear. The skipper made his way forward to the wheelhouse. By this time, the vessel was listing so severely that the starboard side of its superstructure was touching the surface of the sea. Taking hold of the boy, he made two attempts to lift him through the combined hinged hatch cover and door leading outside to the top of the main deckhouse. However, the boy baulked each time because of his proximity to the heated surface of the main engine's exterior exhaust piping. After these two attempts, the skipper placed the boy in the care of one of the crew members, then made his way up through that opening with the intention of boarding and releasing the seine skiff, so that it might be used as a rescue craft for those on board the vessel. He made his way aft along the side of the vessel toward the skiff, which had floated forward when the vessel's listed condition allowed the skiff's painter to slacken. He boarded the skiff and, while attempting to release it, the CapRougeII capsized. The skipper dove underwater at two locations along the hull of the overturned vessel in an attempt to reach those inside. He was unable to do so and, in an exhausted state, managed to re-board the skiff. From this position, he threw a length of rope to one crew member who was in the water and wearing a floater jacket. The skipper hauled the crew member on board the skiff and discovered that, like himself, the crew member had escaped through the wheelhouse top hatch. Three crew members and two children remained inside the capsized vessel and drowned. 1.4 Events Following the Capsizing At 0902, the sailing vessel FrancisLynnII informed the Marine Communications and Traffic Services Centre in Victoria (Victoria MCTS) that a fishing vessel had capsized off Sand Heads Light, which marks the western edge of the Steveston jetty. The FrancisLynnII reported it was approximately two nautical miles south of the position where the CapRougeII capsized, was under sail and unable to proceed directly to assist. The mate on board the sailing vessel observed two persons in the seine skiff of the CapRougeII. No distress call was broadcast by the CapRougeII. No other information about the accident was provided by the sailing vessel. The Ro-Ro passenger ferry Queen of New Westminster overheard the radio communication between FrancisLynnII and Victoria MCTS and informed Victoria MCTS they were off Sand Heads and proceeding to provide assistance. Victoria MCTS issued a MAYDAY RELAY broadcast and the Joint Rescue Co-ordination Centre in Victoria (JRCC Victoria) was advised of the accident. Numerous vessels responded to the broadcast. The JRCC Victoria subsequently tasked dedicated search and rescue (SAR) resources and other vessels to the area. At 0903, the Canadian Coast Guard (CCG) hovercraft CCGH045 departed its base at Sea Island after hearing the FrancisLynnII call Victoria MCTS. On board CCGH045 was the usual crew of foura captain, a first officer and two rescue specialists and another captain, who happened to be on hand when the crew overheard the radio communications about the capsizing. The CCG auxiliary vessel Auxiliary07 in Steveston was tasked by JRCC Victoria. At the same time, 0903, Canadian Forces (CF) 442 Squadron, which is based at Canadian Forces Base (CFB) Comox and on two-hour standby, was informed of the capsizing by JRCC Victoria. A CF CH113 Labrador helicopter was tasked to provide assistance. At approximately the same time, the fishing vessel Taaska was in the area, proceeding toward the capsized vessel. At 0913, the Seaspan Greg also observed the skiff and advised Victoria MCTS that the position of the capsized vessel was approximately 4904'39N, 123 18' 48W. The reported position placed the CapRougeII approximately 1.75nautical miles south of Sand Heads Light. As the Taaska came alongside the seine skiff, her crew was informed by those on board the skiff that five persons were still within the capsized vessel. The Taaska reported this information to Victoria MCTS at 0918. At the same time, 0918, as it approached the accident site, CCGH045 requested of JRCC Victoria that it dispatch a penetration-capable dive team. Within the following minute, CCGH045 had arrived on scene and, assuming the role of on-scene co-ordinator (OSC), began to simultaneously co-ordinate rescue activities and dive operations. At 0920, JRCC Victoria asked Victoria MCTS to inform CCGH045 that the CFCH113 Labrador had been tasked and its estimated time for arrival was two hours. At 0921, CCGH045 contacted Victoria MCTS and requested that JRCC Victoria dispatch a floating crane to assist with recovery operations. Rescue specialists on board the CCGH045 were also preparing to deploy to survey the situation. By 0922, the Queen of New Westminster had arrived on scene and had lowered her rescue boat into the water. The Queen of New Westminster positioned herself upwind from the capsized vessel to provide a lee which would shelter those involved in search and rescue activities. At approximately 0925, the primary diver and standby diver from CCGH045 first entered the water. At 0930, Auxiliary07 arrived on scene and two of its crew members clambered onto the capsized vessel and pounded along the hull but there was no response from inside the vessel. Rescue personnel also tried to detach the skiff from the capsized vessel but were unable to cut the skiff's towing bridle. At 0935, the primary diver began to conduct an underwater external search of the CapRougeII while the standby diver remained in the water but at the surface. During dive operations, the primary diver communicated to the standby diver that he was entangled. The standby diver dove and freed the primary diver. Before either diver was able to resurface, the standby diver became entangled but was freed by the primary diver. The two survivors were removed from the skiff and transferred to the Queen of New Westminster (later, they would be transferred again to the Royal Canadian Mounted Police (RCMP) vessel Nadon). The skipper of the CapRougeII sketched a layout of the vessel for rescue personnel. He informed them that there was one door in the starboard side of the aluminum plating of the wheelhouse but none in the port side. He also informed them of the possible whereabouts of the persons still within the vessel. At approximately 0954, the body of an adult was recovered from within the submerged wheelhouse by CCG divers stationed outside the vessel using a boat hook. A crew member from Auxiliary07 boarded CCGH045 to assist bringing the victim on board. Cardiopulmonary resuscitation (CPR) was then performed. CCGH045 called the Queen of New Westminster and asked if there was a doctor on boardthere was none. CCG divers were unable to perform another dive because their diving cylinders were out of air and there were no spare cylinders on board CCGH045. No other diving cylinders were available on scene. At 0957, CCGH045 issued a MAYDAY RELAY requesting any vessel within the vicinity to provide SCUBA (self-contained underwater breathing apparatus) tanks. Several vessels with charged cylinders on board responded to the call and proceeded to the site. Extra diving equipment, cylinders, and a CAPSAV4 kit from the base at Sea Island were also being brought to the site. At 1003, the CCG cutter Osprey 1informed Victoria MCTS that it had departed its base at Kitsilano and was en route to the site with one diver on board and two spare tanks. At 1004, CCGH045 departed the scene to transport the recovered body, while the crew performed CPR, to an ambulance standing by in Steveston. The hovercraft retained the role of OSC during the 36 minutes it took to reach Steveston and return to the accident site. Auxiliary07 remained on scene and continued efforts to try to release the skiff from the capsized vessel. Other vessels were searching the debris field southwest of the capsized vessel in search of any survivors. A CF CH113 Labrador helicopter, with four search and rescue technicians (SAR TECHs), departed CFB Comox at 0955. Additional arrangements had also been made by JRCC Victoria to transport four divers from the CF Fleet Diving Unit (FDU) at CFB Esquimalt by CCG helicopter to the Sand Heads jetty. Two RCMP divers were also proceeding to the CCG base at Sea Island. CCGH045 was back on scene at 1040, after having transferred the person recovered from CapRougeII to an ambulance in Steveston. The Nadon and Sea Island 1, a CCG rigid hull inflatable craft based at Sea Island, had also arrived. Sea Island 1 had on-board dive cylinders and a CAPSAV kit. After arriving on scene at 1045, the CF CH113 Labrador helicopter lowered the SAR TECHs onto CCGH045. In accordance with CCG operating procedures, CCGH045 had shut down her engines to receive the SAR TECHs. The hovercraft had difficulty restarting the engines and experienced on-board systems failures. At approximately 1100, the FDU divers arrived on scene. The SAR TECHs, CCG and FDU divers transferred to Sea Island 1 to begin their dive operations. CCGH045 departed the scene and returned to its base at Sea Island, so that the crew could transfer to the hovercraft CCGH Siyay. The SAR TECHs entered the water at 1110, with CCG, FDU, and RCMP divers providing backup. Within 10minutes, a child was recovered from within the vessel. The child was then taken by Labrador helicopter to an ambulance at Sea Island. Shortly afterwards, the bodies of two adults were recovered from within the vessel. A further search of the interior of the vessel was conducted. However, the crew's forward, below-decks sleeping area and skipper's cabin could not be searched because of floating debris which prevented divers from reaching those areas. At approximately 1155, CCGH Siyay had arrived on scene and took up the OSC duties. At 1230, dive operations into the vessel were suspended when it was determined by those on scene that the combined factors of the overturned vessel, its suspended equipment and associated debris constituted an undue risk to the safety of the divers. At 1303, a barge with a crane arrived on scene. The crane was unable to lift the capsized vessel out of the water but was able to stabilize the vessel which was subsequently towed to shallow water. Shortly after 1900 on August13, divers were able to search the captain's cabin and recover the body of the second child who had been in the forward accommodation space. None of the persons recovered was located in a pocket of air within the vessel. 1.5 Injuries to Persons 1.6 Damage to the Environment An insignificant amount of diesel was released at the time of the accident. No environmental damage was reported. 1.7 Certification 1.7.1 Vessel Certification As a small fishing vessel of 47 gross registered tons (GRT) and less than 24m in length, the CapRougeII was subject to regulatory inspection on a quadrennial basis. Since being built, the vessel had been inspected in accordance with the Small Fishing Vessel Inspection Regulations(SFVIR). Inspections were routinely carried out at the scheduled four-year intervals and, on 13 June 1994, a Safety Inspection Certificate (SIC)29 for operation as a Home Trade ClassII Vessel was issued, valid until 28 May 1998. Because of operational and/or other logistical reasons, the validity of the SIC29 was extended until 04August 1999 when, following an inspection by Transport Canada (TC), the current SIC29 was issued and made valid until 13September 2002, effectively restoring the original scheduled quadrennial inspection sequence. As part of TC's Ship Registration Program, a Certificate of Registry, with a covering letter, is sent to the registered owner/authorized representative. The cover letter states if your address changes at any time or the vessel is altered, you must notify the nearest Registrar of Ships. The Certificate of Registry has a three-year expiry date from the day the certificate was issued. This new process began in February 2000, as part of the reform of the Canada Shipping Act, and the letter was sent to the owner of CapRougeII on 14 June 2001. 1.7.2 Personnel Certification The skipper of the CapRougeII held a Transport Canada Fishing Master ClassIV certificate of competency, issued 02February 2001, which qualified him to command fishing vessels of this size and type. He received training in marine emergency duties (MED A1) as a mandatory component of course work leading to the issuance of his certificate. A mandatory component of his certificate required him to pass an oral examination on the subject of General Seamanship' - an umbrella category which covered six topics, including Stability'. There is no information to indicate that the four other crew members on board the vessel had received MED training. There is no information to indicate that regular emergency drills including Ship Abandonment' took place on board the vessel. 1.8 Personnel History The skipper had 31 years' experience in the fishing industry and continues to be the proprietor of a company which routinely operates small boats in sorting logs in preparation for their delivery to various mills. All four crew members were experienced fishers and had worked on board fishing vessels for a number of years. 1.9 Weather, Tide, and Current Conditions 1.9.1 Weather Recorded Observations made by the Pacific Weather Centre of Environment Canada at 0900 at the entrance to the south arm of the Fraser River and within two miles of the accident site indicate a wind blowing from the north-northwest (340) at a speed of 17knots. The skies were clear. The waves in the area of the accident were proceeding in a number of directions simultaneously, resulting in what is commonly known as a confused sea. The height of the waves was estimated by the master of a nearby B.C. ferry to be between 1 and 1.5m. 1.9.2 Tide and Current At the time of the accident, the tide was ebbing at a rate of less than two knots after reaching a normal highwater height of 3.8m. At 0833 at the entrance to the south arm of the Fraser River, the downstream current was flowing at less than one knot. 1.10 Emergency Preparedness 1.10.1 Lifejackets and Immersion Suits Life-saving equipment on board the vessel consisted of eight adult lifejackets, two children's lifejackets and five immersion suits. The lifejackets and immersion suits were stowed in the companionway connecting the crew's quarters forward with the wheelhouse. 1.10.2 Inflatable Liferaft The vessel was supplied with a six-person B - type liferaft, manufactured by Beaufort in April 1974. It was serviced on 28February 2001. The liferaft cannister was stowed in a metal, open-topped cradle, located on the command bridge of the vessel. Soon after the vessel capsized, the cannister was observed unopened and floating free of the vessel but still tethered to it by means of the liferaft's painter. 1.10.3 Marine Emergency Duties Training Program As part of Transport Canada Marine Safety (TCMS) requirements to acquire certification as Fishing Master - ClassIV, the skipper successfully completed 191/2hours of instruction on recognizing and responding to hazards in a marine environment. This training program is known as Marine Emergency Duties CourseA1 (MED A1). The syllabus specifies instruction on raising alarms in emergency situations, abandoning a vessel, and the importance of regularly carrying out emergency drills and training. Four of the crew on board at the time of this occurrence participated in a Workers' Compensation Board of British Columbia (WCB) emergency drills workshop, held on Galiano Island, on 11May 1998. 1.11 Vessel Stability 1.11.1 Regulatory Stability Requirements When the CapRougeII was completed in 1974, there were no regulatory requirements calling for the submission and approval of stability data for small fishing vessels. Subsequent requirements were introduced which called for certain small fishing vessels, built after 06July 1977, to submit intact stability data for approval by the then CCG Ship Safety Branch1. (SSB)5. The SFVIR require that only those small fishing vessels engaged in fishing for herring or capelin submit trim and stability data for approval. Consequently, small fishing vessels, such as the CapRougeII, which are principally engaged in salmon seining, are not required to submit trim and stability data for approval, nor is there any mandatory requirement for the owners of these vessels to forward this data to TC for safety review or information purposes. As was the case with the CapRougeII, there is no regulatory requirement for an owner to assess the stability characteristics of a small fishing vessel. However, such an assessment is occasionally done voluntarily because the result is considered an important indication of a vessel's seaworthiness. Regardless of whether a stability assessment is done on a mandatory or voluntary basis, the criteria used in its determination are included in section STAB 4 of the Stability, Subdivision and Load Line Standards (TP 7301). The WCB Occupational Health and Safety Regulation (OHS), Section 24.72(b), requires documentation on board which describes vessel characteristics, including stability. Policy Item R24.72-1, expands upon this requirement stating: Under section 24.72(b), the owner must give notice of unique features of the vessel which might not otherwise be known to a new master and crew and which might cause hazards in certain situations if the boat is not properly handled. This includes instructions on how to perform operations on the vessel without impairing its stability and seaworthiness. However, for vessels such as CapRougeII, the WCB regulation does not define acceptable stability criteria for the guidance of the owner. 1.11.2 Stability Data Approval History The builder of the CapRougeII is no longer in operation; there are no records available of an inclining experiment conducted to verify the vessel's original as-built trim and intact stability characteristics when it was completed as a seiner in 1974. However, a simple rolling period test was carried out at that time, which provided some indication of initial stability, and a transverse metacentric height (GMt) of 19.5inches (495mm) was recorded by the SSB. The actual loading condition of the vessel at the time of the rolling period test is not known, however the derived GMt was considered satisfactory as it was comparable to other small fishing vessels of a similar size and service. In 1987, the vessel was converted for operation as a herring packer by a former owner and, to determine her new stability characteristics, an inclining experiment was conducted on 28December 1987. At that time, all seine nets, the seine net drum and other fishing gear which were not required for herring packing operations were removed from the main deck of the vessel. The original single fish hold was subdivided by new transverse and centreline bulkheads .to provide four separate fish holds. The centreline bulkhead was installed for the full depth of the fish holds, but did not extend up to the top of the main deck hatch coaming in way of the forward pair of holds. The trim and stability data derived from the results of the experiment showed the vessel exceeded the minimum stability criteria of STAB4, which are applicable to small fishing vessels engaged in catching herring or capelin. However, the requirements were not applicable to herring packing operations, and the inclining experiment was not witnessed by a TC inspector. The related trim and stability booklet was not submitted to or approved by TC. 1.11.3 Vessel Modifications History Since being built in 1974, the vessel has been the subject of several modifications and the installation of additional rigging, fishing, hull, and engine room outfit. Previous and present owners have added to, and modified, the outfit of fishing gear and other equipment to meet the ongoing changes, developments and requirements of the seining fishery. Since 1987, when the vessel was arranged for fish packing operations with fishing gear and the powered seine net drum removed, the CapRougeII had been re-rigged and updated for current salmon seining operations by the modification of existing equipment and the installation of various items of additional gear, including the following: a full-width stern seining ramp, roller, hydraulic package, and controls; seine net power drum seating raised some 330 mm to increase under-net clearance; a fish pen permanent coaming fitted on main deck in way of hold loading manholes; a 2.44m-diameter seine net power drum, hydraulic drive package, and controls; main cargo boom vanging winches, rigging, hydraulic package, and controls; main cargo boom topping winches, rigging, hydraulic package, and controls; hydraulic seine net power block on main boom; Volvo auxiliary engine and electrical generator set in engine room(S); bow thruster unit installed in compartment within forward freshwater tank; increased hydraulic and lubricating oil tank capacities in engine room; seine net ring stripper davit at top of bulwark rails near midships(S); heavy-duty domestic deep freezer on deckhouse top; additional radar mast, unit, and scanner on wheelhouse top; additional fishing lights on main mast; permanent trimming ballast on main deck in captain's cabin (430kg)(S); accumulated spares and tools in engine room (570kg); and West Coastseine net on power drum at 1.73m above main deck (7400kg). The cumulative effect of the above items increased the vessel's lightship weight, lowered the inherent effective freeboard, raised the centre of gravity and markedly reduced the transverse stability characteristics of the CapRougeII. 1.11.4 Vessel on Recovery During and after the salvage and recovery of the CapRougeII, the vessel was inspected by divers, the Transportation Safety Board of Canada, TC, and WCB personnel. The following items were noted: the seine net was on the net drum. (Dry weight was subsequently verified to be 7400kg); the main boom was in the raised position and secured on the centreline of the vessel; the roll reduction outrigger booms (stabilizers) were secured in the vertical position; four main deck flush-fitting manhole covers, giving access to the fish holds and another in way of the lazarette, were found in place by the divers, but were subsequently removed to facilitate pumping the vessel clear of floodwater during the salvage operations; the door on the starboard side of the wheelhouse was found closed by divers; the combined hinged hatch cover and door from the after end of the wheelhouse, giving access to the main deckhouse top, was found open by divers; the door in the after end of the main deckhouse, giving access to the after end of the main deck, was found secured in the open position; the fibreglass covers of the main deck hatch and fish sorting table were missing; unsecured spare parts and tools, displaced from the top of the port side oil fuel tank, were found on the starboard side of the engine room; autopilot controls in the wheelhouse were found in the engaged position; the rudder was found to be turned approximately 20 to starboard; the valves in the suction and overboard discharge piping systems of the after starboard side fish hold were open and the fish hold pump set for discharging water from that compartment; valves in the suction and return pipes cross connecting the two after oil fuel tanks were open and the oil fuel transfer pumping system set for the tanks to be used simultaneously; valves in the suction and return pipes cross connecting the oil fuel tanks in the engine room were all closed, and the port and starboard tanks were isolated from each other; the oil fuel pumping and transfer system was set to draw fuel from the after tanks only; when the after oil fuel tanks were emptied and their contents measured by a calibrated flow meter, they were each found to contain 2273L (500gallons); the oil fuel tanks in the engine room were emptied and their contents were also measured by flow meter; the starboard side tank contained 2273L (500gallons) of diesel oil, and the port side tank held 1727L (380gallons); the lazarette was found to contain 1410L (310gallons) of seawater; the hydraulic oil storage tank was found to contain 1136L (250gallons) of oil; and the forward freshwater tank contained 1363L (300gallons). Hose testing and inspection of the flush-fitting manholes in the main deck, giving direct access to the four fish holds and the lazarette, showed that their flexible sealing gaskets were ineffective and not watertight. During the hose tests, some water leaked into the forward pair of fish holds, past the gaskets of the secured 610x380mm (24x15inches) manhole covers. A significant amount of water entered the after holds past the gaskets of the 940x610mm (37x24inches) secured manhole covers. A steady stream of water was also found to enter the lazarette past the gasket of the 610x455mm (24x15inches) secured access manhole cover. Water entered each of the after fish holds in significant quantities, with the port side being somewhat greater. However, the manhole covers were identical in appearance and, because they had been removed during the salvage operations, their correct re-location at the time of the tests could not be verified, and the actual athwartship imbalance of the downflooding rates into the port and starboard after holds at the time of the capsize could not be determined. As found on recovery, the open or closed condition of the wheelhouse and main deckhouse doors of the vessel were consistent with the sequence of events prior to the capsize, as reported by the survivors. The secured condition of the lazarette and fish holds flush-fitting manhole covers and the configuration of the oil fuel transfer piping and water ballast pump operating arrangements were also as reported. 1.11.5 Occurrence Stability An inclining experiment was carried out by the TSB on 26 August 2002 to determine the lightship condition of the CapRougeII at the time of the capsize and to assess the effects due to the weight of additional fishing gear and other equipment primarily installed at or above main deck level. A rolling period test was also carried out at this time, which provided on-site confirmation of the as inclined initial stability condition of the vessel. Both of these tests6 were attended and witnessed by a Minister of Transport observer and a TC inspector. Preparations for this inclining experiment revealed that the existing after draught markings on the hull were set some 112mm (4 inches) too high, and that those forward were also approximately 12mm (1/2inch) too high. Consequently, the lightship weight derived from the previous inclining experiment 28December 1987, was significantly lower than that of the actual vessel at that time, making it and the related stability characteristics unreliable for comparison purposes. An assessment of the intact transverse stability was carried out to verify the condition of the CapRougeII prior to and at the time of the capsize, based on the current lightship data, the reported fish catch, the recovered on-board consumables, the West Coast seine net and fishing gear, etc. Calculations show that, shortly before the capsize, the vessel had an after trim of approximately 0.80m (2.67ft). The initial transverse stability in the static condition, as indicated by the GMt, exceeded the minimum STAB4 criteria by some 23percent. However, the maximum righting lever (GZ) was 52percent of the recommended minimum and the total dynamic stability, as represented by the area under the righting lever curve, was 50percent of that required up to an angle of heel of 40. The maximum righting lever was reached at 20 heel in lieu of the minimum recommended 25 and the range of positive stability was limited to 40 angle of heel. Due to the wave pattern generated by the speed of the vessel through the water and the rolling and pitching motions in the prevailing confused sea conditions, water was shipped through the freeing ports at main deck level and also through openings in the fixed stern ramp. As the shipped water accumulated at the after end of the main deck, it was constrained within the area bounded by the seine net drum seating and the forward fish pen and hatch coaming. Because the gaskets of the manhole covers of the four fish holds and the lazarette were ineffective, downflooding into all of these compartments commenced (see Figure3). The accumulated water on deck, together with the downflooding of the lazarette and lightly loaded after holds, created a marked increase in free surface area and a corresponding reduction in transverse stability. The additional weight and free surface effects of the shipped water increased the after trim to 1.22m (4.0ft), lowered the mean and after freeboards, and markedly lowered the transverse stability characteristics of the vessel. While in this condition, the GMt was reduced to 68percent of the minimum requirement and, more significantly, the maximum righting lever and dynamic stability were lowered to 19percent and 10percent, respectively, of the recommended minimum values. Furthermore, the range of positive transverse stability was limited to an angle of 22 in lieu of the recommended range in excess of 40. Figure 3. Downflooding prior to capsize 1.11.6 Heeling Capsize Sequence Reportedly, the vessel routinely maintained a small angle of heel to starboard and was regarded as a wet ship, which regularly shipped water on deck when loaded. This inherent heeled condition is consistent with the recorded imbalance of the diesel in the engine room tanks, the asymmetric volume of the freshwater tank following the installation of the bow thruster, and the location of the solid trimming ballast in the captain's cabin on the starboard side of the vessel. Because of this initial heel, slightly more seawater was shipped and retained on the starboard side of the vessel, and the diesel in the cross-connected after fuel tanks was free to gravitate to starboard. The rate of downflooding into the port and starboard after fish holds differed due to the uneven condition of the manhole cover gaskets. The uneven downflooding rate is most likely due to water predominantly entering the starboard after hold, which increased the heel to that side, and first alerted the crew to the unusual attitude of the vessel (see Figure3). The calculated transverse stability characteristics show the CapRougeII was in a highly vulnerable condition in the confused sea conditions. With the winds and waves acting primarily on the port side, the dynamic ship motions and free surface effects of the shipped water and below deck liquids caused large fluctuations in transverse stability. In these circumstances, the sudden influence of a single, or combination of relatively small, additional heeling forces would cause the loss of the remaining transverse stability and lead to capsize. The reported sudden development of the large heel to starboard is a typical reaction to the loss of transverse stability, due to large free surface effects. In this instance, recovery was initially retarded and then prevented by the sudden gravitation of the liquid contents of all under-deck compartments and an unsecured engine room outfit to the lower side of the vessel. Because the centreline bulkhead between the forward pair of fish holds did not extend to the top of their common hatchway, the contents of the partially filled port side fish hold were free to flow to starboard. This large weight transfer prevented the vessel's recovery. The gravitation of diesel from the port to starboard after fuel tanks, by way of their cross-connection piping, also contributed to preventing any recovery. In the confused seas immediately prior to the capsize, when heavily trimmed by the stern and with a large heel to starboard, the CapRougeII was in an extremely vulnerable condition. At this time, a combination of relatively small additional heeling forces overcame the vessel's remaining righting ability and completed the capsize sequence. This combination included the yawing action of the downward force of the stretched tow rope of the seine skiff where it was led over the top of the seine net drum (see Figure3), and the upward force of a wave acting on the port underside of the fixed stern ramp. 1.12 History of CCG Rescue Dive Team The impetus for the provision of rescue dive services on the west coast dates back to 1992 in Nanaimo, when a vehicle, with passengers on board, fell from a ramp used in loading the Queen of New Westminster (TSB ReportM92W1057). The accident prompted a provincial commission of inquiry into the resulting deaths. Among other issues, the inquiry, led by Mr. Justice Nemetz, considered the provision of rescue dive services, concluding that the issue was beyond the scope of the inquiry and recommended a provincial task force be set up within six months to consider the provision of rescue dive services. The resulting task force was unable to come to a definitive conclusion with respect to the issue of rescue diving, although it acknowledged that the probability of divers being able to rescue underwater victims was extremely low. The issue of rescue dive services came into the public eye again in 1993, following a collision between the vessels Bona Vista and Arctic Taglu (TSB ReportM93W1050) which resulted in the capsize of the Bona Vista. A number of people survived for several hours in an air pocket within the overturned hull but were not rescued in time. In response, CCG set up a working group in January 1994 to consider the provision of rescue dive services. The working group was unable to reach any definite conclusions and, in June 1995, a Rescue Diving Pilot Project became operational. The project was to run for a period of two years to allow the feasibility and benefits of rescue diving to be assessed. The pilot project involved training and equipping volunteer divers at CCG Station Sea Island to operate in conjunction with the hovercraft operation. The volunteers were drawn from existing members of the hovercraft crew and duties were carried out in addition to an individual's normal duties as a hovercraft captain, first officer, or rescue specialist. It was attempted, wherever possible, to schedule crews such that the hovercraft was always staffed with two qualified divers and a qualified dive tender. At the end of the two-year period, in 1997, a review of the pilot project was conducted7. The project had not provided sufficient data to allow a definite conclusion to be reached. It recommended the pilot project be extended for an additional two years to allow a full evaluation of the benefits. During this two-year extension, on 15 September 1998, the Canadian Occupational Health and Safety Regulations (COHS) - Canada Labour Code (CLC), Part II took effect. The Code, applicable to CCG dive operations, states that: 18.48Where there is a likelihood that a diver may be entrapped, the employer shall ensure that a two-way voice communication system between the diver and the diver's tender is provided; and a second dive team, equipped to rescue a diver in the event of an emergency, is present at the dive site. Although the CCG rescue dive team was already equipped with a two-way voice communication system, the requirement for a second dive team to be present at the site raised questions for the team, whose work involved entering overturned vessels, aircraft, and vehicles. Following the implementation of these Regulations, a conference call took place between the Officer-in-Charge (OIC) of the dive team, the Regional Dive Officer and the National Dive Officer, where they agreed that having a second team en route to the dive site prior to commencing a dive where a danger of entrapment would exist would be acceptable. Subsequently, the four-year review of the pilot program was prepared by the OIC of the CCG dive team in 20008. It concluded that the team should become permanent and be granted an appropriate level of support for a permanent SAR program. Later that same year, the Commissioner of the CCG asked the Regional Director of the Pacific Region to look into the issue of whether CCG should be in the business of rescue diving. In response to this request, a third review of the rescue diving project was initiated. This report9 identified a number of problems in the operation of the dive team and concluded that there were two possible scenarios for rectifying the deficiencies: the dive program should either be cancelled or sufficient resources should be devoted to allow it to be conducted safely. As a follow-up to this report, the dive project was suspended on 29November 2000 pending further review. This review10 was conducted by a designated CCG officer and an external consultant. The report, submitted on 25January 2001, found that there were deficiencies in the operation of the dive team which could be rectified. Most deficiencies concerned documentation surrounding the training of dive personnel and the provisions in place for the availability of a second dive team to comply with the CLC requirements for penetration dives. On Friday, 16February 2001, the Regional Director of Pacific Region met with members of the dive team at the Sea Island Hovercraft Base to announce that the dive project had been terminated. Diving equipment was immediately removed from the hovercraft. Later that weekend, in the early morning hours of Sunday, February18, a car accident occurred in which a vehicle entered the water in close proximity to the Sea Island base. The hovercraft reached the scene quickly and it was anticipated by other emergency response personnel on scene that the Coast Guard would conduct a rescue dive, as they were unaware the dive project had been cancelled 36 hours previously. The driver of the car did not survive the accident and the Minister of Fisheries and Oceans received significant public attention surrounding the cancellation of the rescue diving project. This attention prompted a review of the decision-making process respecting the project11. The review concluded that the decision to suspend and ultimately cancel the dive project: was correct, even if made within a rather faulty framework. The report further concluded that: the attitude of senior CCG management towards rescue diving will probably not be changed unless the Minister directs them to establish a policy, create a national program and generally support the rescue diving concept both philosophically and financially. Following this review, the project was reinstated as an Open Water Rescue Dive Team' and the Sea Island base set about training rescue divers. By September 2001, the base was dive capable on approximately 60percent of shifts and, by January 2002, this had increased to being dive capable on a continuous basis. During this period, a series of documents defined the open water' limitation of the dive team. The CCG SAR Station Sea Island Diving Manual, dated 20August 2001, restricted the team to open water dives but provided for the continued training of divers to enter capsized vessels: The Pacific Region version of the CCG Fleet Safety Manual (FSM), dated 15July 2002, further clarified the issue of divers conducting penetration dives: In keeping with these directives, CCG divers were prohibited from entering the overturned hull of the CapRougeII on 13 August 2002. Following this occurrence, the Minister's office made a statement which indicated the divers could conduct penetration dives, in that provisions in the Canada Shipping Actdelegate the power of the Minister to the SAR co-ordinator at the Rescue Coordination Centre. As such, a decision to conduct a penetration dive could be made by the SAR co-ordinator in conjunction with the CCG vessel captain, possibly without a backup dive team in place. It was further stated that this provision was in effect at the time of the capsize of the CapRougeII. Modifications to the FSM, effective 08November 2002, outlined the conditions: where the craft captain can request a direction from the SAR Mission Co-ordinator (JRCC/MRSC) to attempt to enter the vessel, aircraft, or vehicle including: the dive being necessary to save a life, the concurrent dispatch of a backup dive team and the use of surface-supplied breathing apparatus once this equipment was available (in the interim, penetration dives using SCUBA would be permitted). On 06 February 2003, the Minister of Fisheries and Oceans announced an expansion of CCG rescue dive capabilities at the Sea Island base. Included in this announcement were provisions to expand the team to allow a dive team with full backup to be available 24hours a day. In addition, it was indicated that the dive team would be equipped to allow the use of surface-supplied air. 1.13 Responsibilities of On-scene Co-ordinator and Dive Operations The mission of the CCG SAR program is to save and protect lives in the maritime environment. One of the CCG's objectives is to maintain the highest professional standards. CCG management's goal is to ensure that the SAR program operates at maximum efficiency by adjusting SAR coverage requirements as needs change, and by deploying specialized primary SAR units as required. Additionally, other resources (private and government) may be tasked to provide assistance during a SAR operation. Those resources may include vessels, helicopters, and aircraft, which may happen to be nearby or capable of providing specialized services. It is not uncommon for a number of resources to be involved simultaneously in a SAR operation. The JRCC Victoria SAR Operations Report for this accident13 listed 17resources which were tasked to respond. Consequently, there is a need for co-ordination of on-scene SAR activities. In Canada, a JRCC is established within each of Canada's search and rescue regions for the purpose of co-ordinating, controlling, and conducting air and maritime search and rescue operations. In respect of maritime operations14, JRCC staff responsibilities include planning and directing the response, tasking and co-ordinating resources, and appointing an OSC when appropriate. An OSC is normally appointed to co-ordinate SAR operations where several resources have been tasked to respond. An OSC is responsible for the following tasks, to the extent they have not been performed by the responsible JRCC or Maritime Rescue Subcentre (MRSC) as listed in the SAR Seamanship Reference Manual:15 carry out the plan for the conduct of operations as requested by the responsible [J]RCC/MRSC; modify the plan as facilities and on-scene conditions dictate and inform the [J]RCC/MRSC of any such modifications; monitor weather and sea conditions and report on these at regular intervals to the [J]RCC/MRSC; maintain communications with the [J]RCC/MRSC and the SAR units on scene; maintain a detailed record of the operation, including on-scene arrival and departure times of SAR units, areas searched, track spacing used, sightings and leads reported, actions taken and results obtained; issue regular situation reports to the [J]RCC/MRSC, which should include, but not be limited to, weather and sea conditions, the results of search to date, any actions taken, and any future plans or recommendations; and advise [J]RCC/MRSC to release units when their assistance is no longer required. CCG Pacific Region SAR services also include rescue dive services, which are provided by the Sea Island Hovercraft Unit located in Georgia Strait. The officers and crew of CCGH045 were designated to provide rescue dive services and, in this accident, were engaged in rescue dive operations. Typically, the safety of any non-recreational dive operation requires that the following duties be undertaken in support of a primary diver engaged in underwater operations: the operator of the dive platform (i.e., the vessel) remains devoted exclusively to the control of the vessel during live boating;16 a dive supervisor is devoted exclusively to the dive operation; a standby diver stands ready. CCG standard operating procedures for rescue dive operations also require that the rescue unit go through pre-arrival, on-scene and diver preparation procedures prior to diving. There are also procedures to be followed during a dive. CCGH045, with a crew of five, was on scene within 20minutes of the distress call. Standard operating procedures called for two officers and two rescue dive specialists to crew the hovercraft, but, on this occasion, an additional officer was on board.